The nucleus basalis of male Charles River Wistar rats was injected with 10 micrograms of the beta-amyloid peptides beta-(1-40) and beta-(25-35) and changes in the morphology of the lesioned area, the release of acetylcholine from the cortex, and in behavior were investigated. Injections of saline and a scrambled (25-35) peptide were used as controls. One week after lesioning, a Congo Red-positive deposit of aggregated material was found at the beta-peptides injection site, which lasted for about 21 days in the case of the beta-(25-35) peptide and at least two months for beta-(1-40). No deposit was detected after scrambled peptide injection. At one week post injection, an extensive glial reaction surrounded the injection site of all peptides and saline as well. Such a reaction was still present but rather attenuated after two months. A decrease in the number of cholinergic neurons was detected in the nucleus basalis after one week with all treatments except saline. After two months, a reduction in the number of choline acetyltransferase-immunopositive neurons was still detectable in the rats injected with beta-(1-40) but not in the beta-(25-35)-or scrambled-injected. The reduction in choline acetyltransferase immunoreactivity was closely paralleled by a decrease in basal acetylcholine release from the parietal cortex ipsilateral to the lesion. Disruption of object recognition was observed in the first weeks after beta-(25-35) peptide injection, whereas the beta-(1-40) peptide impaired the performance only two months after lesion. Rats with lesions induced by beta-peptides may be a useful animal model of amyloid deposition for investigation of the pathogenetic mechanisms leading to Alzheimer's disease.